AI Post — Artificial Intelligence

🔥Milestone in Humanoid Robotics: A thousand humanoid sorters entering logistics centers Beijing-based RobotEra is deploying its L7 humanoid robot across more than 10 logistics centers operated by China Post, SF Express Group, and other major players. In several of these centers, the embodied AI robots have already reached over 85% of human-level efficiency while operating stably 24/7. The company is set to begin batch deliveries of robots at the thousand-unit scale in Q2 this year. RobotEra recently raised $200 million in funding. By combining external capital with self-generated revenue, it is accelerating the real-world deployment of humanoid robots. @aipost🏴

"OpenAI has a 50% chance of going bust in the next 18 months." Sebastian Mallaby explains why: • Spread too thin • Burning cash at a historic rate • Fundraising questions Full video. @aipost🏴

📊The biggest IPO run in the history of the market Three $1T+ companies. Possibly all going public in the next 12 months. 🚀 SpaceX IPO: $1.75T. 🤖 OpenAI IPO: $1T 😒 Anthropic IPO: $1T We're living through the greatest technological wealth creation in history. @aipost🏴

⚠️ An AI coding agent powered by Claude just deleted an entire company’s production database in 9 seconds. Cursor, running Anthropic’s Claude Opus 4.6, was assigned a routine task for PocketOS, a SaaS platform serving car rental businesses. Instead of stopping…

⚠️An AI coding agent powered by Claude just deleted an entire company’s production database in 9 seconds. Cursor, running Anthropic’s Claude Opus 4.6, was assigned a routine task for PocketOS, a SaaS platform serving car rental businesses. Instead of stopping when it hit a barrier, the agent made its own decision: Delete the Railway cloud volume. One API call later: • Production database erased • Volume-level backups erased too • Months of customer data gone When questioned afterward, the AI reportedly admitted: “I guessed instead of verifying. I ran a destructive action without being asked. I didn’t understand what I was doing before doing it.” The failure wasn’t just the model. Railway’s setup made it worse: • Backups stored on the same volume as live data • No confirmation prompt for destructive actions • Single command could wipe everything instantly Founder Jer Crane is now rebuilding bookings manually using Stripe payment logs, email receipts, and a 3-month-old backup. @aipost🏴

🗣OpenClaw creator Peter Steinberger talks about how China is going all-in for AI agents and OpenClaw. Full video. @aipost🏴

❗️Here's a rough summary of Elon Musk’s 1 hour and 40 minute long testimony today during the OpenAI trial. He will resume his testimony tomorrow. • Argues the case has huge implications: "It is not ok to steal a charity. If the defendants are found not guilty, this case will become caselaw. It’ll give license to looting every charity in America. The consequences of this case go far beyond me or everyone here. The entire foundation of charitable giving in America will be destroyed." • Says OpenAI was founded as a nonprofit, open-source counterweight to Google, focused on AI safety. • Claims the shift to a for-profit structure violated that mission. • Says his AI concerns date back to conversations with Larry Page, who he felt wasn’t taking AI risk seriously. • Elon says he tried to warn Obama about AI, but that Obama felt AI was not good enough (back then) to seem scary smart. "Here we are in 2026, AI is very smart." • Believes AI could surpass human intelligence as soon as next year and poses existential risk in the hands of the wrong people: “If you have someone who’s not very trustworthy in charge of AI, that’s very dangerous for the whole world.” • Elon framed his companies (SpaceX, Neuralink, xAI) as part of a broader mission to protect humanity’s future. • Emphasized OpenAI’s original goal: AI for the good of humanity, not profit-driven control. • Elon's main argument is that OpenAI abandoned its founding principles, and that precedent could reshape both AI governance and charitable trust. • Larry Page refused to speak to Elon Musk again after Elon recruited Ilya Sutskever to join OpenAI. Elon viewed Ilya as the “number one” most valuable member at Google. • Elon thought in the early days, OpenAI's corporate structure would be a nonprofit funded initially with donations, but there could potentially be a parallel for-profit that is owned by the nonprofit and funds the nonprofit: “We (Sam Altman and Elon) were in agreement that OpenAI would be a 501c3 charity. • Elon was not opposed to there being a small for-profit that provided funding to the nonprofit, as long as "the tail didn’t wag the dog." • Elon: “there are very few people who understand venture capital in Silicon Valley like I do." @aipost🏴

Elon Musk on X. @aipost🏴

🤣Elon Musk's lawyer's microphone was turned off four times during his opening speech in the "Musk vs. OpenAI" case. The federal judge stated, "We are funded by the federal government." The judge also requested no social media activity during the trial. Elon Musk reposted negative news about Sam Altman, with some posts appearing at the top of feeds via paid promotion. Read the court proceedings at CNBC. @aipost🏴

✈️Telegram turns bot creation into a chat prompt Telegram just introduced a native AI-powered “master bot” that can autonomously build, launch, and manage custom bots directly inside the app. No coding. No external tools. No developer setup. How it works: • Users simply describe what they want in natural language. • The AI translates instructions into a working Telegram bot. • Bots can be deployed instantly within Telegram’s ecosystem. • Ongoing edits, upgrades, and management can also be handled through conversation. Now creators, small businesses, communities, and solo operators can launch: • Customer support bots • Lead generation funnels • Community moderators • Content delivery bots • Mini SaaS tools inside chat @aipost🏴

❗️A 23-year-old has cracked one of the Erdős problems that remained unsolved for over 60 years, using ChatGPT 5.4 Pro. Now imagine what ChatGPT 5.5 Pro will be capable of. And remarkably, it was done in a single pass. Total solve time: just 1 hour and 20 minutes. @aipost🏴

⚡️Meta bets on orbital solar power and 100-hour batteries for AI Meta announced two futuristic energy infrastructure partnerships designed to secure power for its growing AI data center footprint while reinforcing the wider grid. Instead of relying only on conventional renewables, Meta is investing in next-generation energy systems that sound closer to science fiction than utility planning. Deal 1: Solar power from space: Meta is partnering with Overview Energy to develop space-based solar power systems. • Goal: beam up to 1 gigawatt of solar energy from orbit to Earth • Generates power 24/7, unaffected by clouds, seasons, or nighttime • Could provide always-on clean electricity for AI workloads and grid demand Deal 2: 100-hour energy storage: Meta is also partnering with Noon Energy to deploy up to 1 GW / 100 GWh of ultra-long-duration storage. • Designed for 100+ hours of continuous discharge • Helps solve renewable intermittency • Stores excess energy and releases it during shortages or peak demand • Could stabilize grids with heavy solar and wind penetration @aipost🏴